Getting Started With the Prior Learning Data

This combination worksheet and mentor session activity is to help everyone get started with the Prior Learning Data.

The prior learning survey can be found here.
You should open this file and peruse it. Think about what questions are interesting to you.
Write down the question numbers that you want to access. Write down the way the answers are recorded. In all cases except major, the answer is a small integer. For example question 110 is gender, the answers will be either 1=male or 2=female, other answers are possible (usually 99=other and 101=left it blank). For each question of interest make a record of what legitimate answers could be.

Down load each of the following into your code directory on your H drive or on your lap top.

Down load the prior leaning stub program to get started.
Be sure you have downloaded the newest versions of Excell.hs and the CSV.hs libraries into the same directory. (Download them and just overwrite the old versions).
You will also need to download the PriorLearn.hs library

The prior learning data is stored in a named declaration in the PriorLearn.hs library named priorLearnData.
Look at the first two elements of the list by typing: take 2 priorLearnData.
It should look like this.
Run an experiment. Use then length function to find out how many people took the survey.
In general we will access data by using comprehensions. Get the gender of all students by typing:
[ gender110 x | x <- priorLearnData ] .
There are lots of answers. So many that the results run off the screen. Here is part of what appears on my screen.

Data is accessed by using selector functions like gender110. Each slector function selects the answer from one question.
There are over 120 different selector functions. You can find the name of a selector function by using the name function in Hugs when the PriorLearn.hs library is loaded.
Try it out, choose a question number, say 23, and type name 23.
I ran a few examples, see my results.
You can get all answers to question 23 by typing [ psu_FriendsAttend23 x | x <- priorLearnData ].
Mostly we want only selected answers. We use the selection feature of comprehensions to do this. For example I may write
```
[ age109 x | x <- priorLearnData
           , frinqRegThemeInterest32 x == 1
           , workHoursThisYear43 x == 4 ]
```
For the ages of all survey participants whose theme interest corresponds to answer 1, and whose work hour corresponds to answer 4. How do I know what values the answer coorespond to. Look them up in the prior leanring document!

I have started to write a program in a stub file. Down load it and load it into Hugs.
What questions does this program deal with?
What are the possible legal answers for each of the selector functions?
Load it into hugs, and explore. What value does temp1 have?
type temp2, what do you get? How can you interpret this data?
temp3 illustrates a very important function mapGroup.
mapGroup id when applied to a list of pairs, and groups all elements together with the same first pair. Each group has the common first element, and a list of all the second elements. Try it by inspecting temp3.
Now try inspecting temp4, which is similar to temp3, except it uses length on all the second elements. How can we interpret temp4?
What is temp5?
Can you use the function highschool to get an interesting answer?
Can you write your own function to use with mapgroup?